Approximately two and a half weeks left of grad school. Fueling up while reading about disseminating research.
“As a matter of fact, ANOVA and the Kruskal-Wallace test are both for situations where groups being compared are not related, the more commonly encountered situation. If the groups are related, then we should use repeated measure ANOVA instead of the ordinary ANOVA discussed in this book, and the Friedman test instead of the Kruskal-Wallis test. These tests are not usually covered in basic statistics books.”
“Many research questions are oriented to discovering whether there is a relationship between two variables and, if there is, how strong that relationship is. Even more to the point, the question is likely to be focused on whether one variable, in some sense, causes the other. Although the statistical methods of correlation cannot directly address this later question, they can provide evidence of a relationship between two variables.”
“. . . no matter what statistical decisions we make about the null hypothesis, there is always a chance of error. This is understandable because we never know the absolute truth of the population-if we knew, there would be no need to do a statistical test. All we are doing in inferential statistics is to estimate the population based on a sample by employing knowledge of probability theory.”
“When using participant observation you should feel confident about answering ‘yes‘ to the following questions:
1 Does the research focus on getting insights to events, cultures and lifestyles?
2 Have measures been taken to minimize any disturbance to the naturalness of the setting?
3 Has consideration been given to the ethics of the fieldwork (secrecy, consent, confidentiality)?”
“The use of evidence gathered directly from preexisting documents or artifacts can greatly strengthen a study. In an appropriate sample can be gathered, this nonreactive approach to data collection can allow the researcher to see some aspects of a situation that could not be detected through more intrusive data collection methods such as interviews or questionnaires.”
“Focus groups can be a good choice for data collection if you believe that value will be obtained in the participants‘ interactions with each other, rather than through individuals responding to questions from you, the researcher. As several of the authors cited in this chapter have argued, people form their opinions in a social/group setting. Focus groups are not an ideal data collection method for every situation . . .”
“When conducting a focus group, take group dynamics into account. Whenever you gather two or more people into a single interview, these individuals will rarely act as true equals. Some participants are likely to dominate the conversation. Others may be reluctant to express their views, perhaps because they‘re shy or feel uncertain about the validity of their perspective.”
Back to the grind - grilled cheese and cafe au lait for lunch while I do some reading for grad school (on surveys)
“Although some progress has been made, there is evidence in the literature that there is a long road ahead of us as a profession. However, the focus on educating the next generation of LIS professionals is encouraging.”
